1. Field of the Invention
The present invention relates to transgenic animal models for human bladder cancer and their use as in vivo models for testing potential carcinogens, preventative measures, as well as therapeutic modalities for intervention in the progression of human bladder cancer.
2. Description of the Related Art
Bladder cancer is the fifth most common cancer affecting predominantly the aging male population and the twelfth leading cause of cancer deaths. Despite aggressive therapy, bladder cancer still continues to be one of the leading causes of cancer deaths in the United States. The American Cancer Society estimated that in 1997 there were 54,500 new cases diagnosed and 11,700 deaths resulting from this disease in the United States.
A major cornerstone in bladder cancer research is the recognition of two distinct forms of bladder cancer (Koss, 1985 and 1992). About 80% of transitional cell carcinomas (TCCs) are superficial, well-differentiated, papillary tumors. This type of TCC is often multifocal and recurrent but rarely progresses to muscle invasion. Another type accounts for 20% of all TCCs and is either present as a carcinoma in situ (CIS) or as poorly differentiated, invasive tumors. It is believed that the majority of invasive TCCs are not derived from the superficial type but directly from CIS and related urothelial abnormalities (Koss, 1992). Interestingly, recent genetic studies suggest that these two forms of bladder cancer may have distinct underlying genetic causes. Cordon-Cardo et al. (1994) proposed that the inactivation of p16 may be responsible for the formation of superficial TCCs, while the inactivation of p53 underlies the invasive form of TCCs.
A set of growth-promoting and growth-inhibiting genes tightly controls cellular growth and differentiation and serves to ensure normal development and to safeguard tissue homeostasis. Alterations in the genes can lead to cellular transformation and tumor formation (Witkowski, 1990; Hunter 1997). Research in the past two decades has identified two classes of genes whose alterations play a major role in tumorigenesis. The first class of genes is oncogenes that were initially identified in studies on retroviruses (Bishop, 1991). These genes have cellular counterparts, proto-oncogenes, that promote normal cell growth; but when activated by a point mutation or induced to overexpress, these genes can promote tumorigenesis in a dominant fashion. The second class of genes is tumor suppressor genes that suppress cell growth, and their mutation or functional inactivation can also contribute to tumorigenesis (Weinberg, 1991). Unlike oncogenes, tumor suppressor genes act in a recessive manner, because the loss of activity requires the inactivation of both alleles. Studies from a variety of human solid tumors suggest that the concerted activity of these two classes of genes underlie tumor development and progression (Finlay, 1993; Vogelstein et al, 1993; Hoskins et al, 1994).
Many genetic alterations involving both oncogenes and tumor suppressor genes have been associated with bladder cancer. Of these, the most notable are the activating mutation of Ha-ras oncogene, the overexpression of neu/erbB-2 and myc oncogenes, and the inactivating mutations of tumor suppressor genes, p53, p16 and retino blastoma gene (Rb) (Borland et al, 1992; Knowles, 1995; Cordon-Cardo et al, 1997) discussed below.
Ha-ras (H-ras): Activating mutations of the ras gene family (Ha-, K- and N-ras) are detected in about 30% of all human tumors, making them the most frequently mutated oncogenes (Bos, 1989; Fearon et al, 1990). Ras mutations found in human tumor and experimental animal models involve predominantly codons 12, 13 and 61. Although ras mutation was first identified in bladder cancer (Capon et al, 1983), its precise role in bladder cancer formation remains unsettled. First, differences in ras mutation frequency has been noted. Earlier functional assays detected ras mutations in 7-16% of bladder cancers (Fujita et al, 1985; Pulciani et al, 1987). More recent data from Koss and colleagues using more sensitive assays detected a much higher rate of up to 40% (Czerniak et al, 1992). Second, it remains unclear whether ras activation can act as an initiating event in bladder cancer formation. Ras mutations are believed to be an initiating event in skin, lung, breast and bowel tumors (Daya-Grosjean et al, 1993; Finlay, 1993; Vogelstein, 1993; Li et al, 1994; Hoskins et al, 1994). In bladder cancer, however, ras mutations have been primarily associated with higher grade, later stage tumors (Levesque et al, 1993). Third, the transformation potential of ras oncogene remains controversial. While activated ras alone can efficiently transform immortal cell lines, it requires the cooperative activity of another oncogene (myc) or a dominant-negative p53 to transform primary rodent cells (Coopersmith et al, 1997; Serrano et al, 1997). On the other hand, the high-level expression of activated ras can circumvent the need for a cooperating oncogene to transform rodent cells (Mann et al, 1991).
Neu/erbB-2: The erb-B2 proto-oncogene encodes a receptor tyrosine kinase with sequence homology to epidermal growth factor (EGF) receptor (Hynes et al, 1994). Amplification of erbB-2 gene was found in 14% of human TCCs overall, but in up to 46% of grade 3 tumors (Coombs et al, 1991). Increased expression of erbB-2 protein (p185) can be detected in an even higher percentage (30-60%) of late stage TCCs, suggesting a correlation between erb-B2 overexpression and tumor progression (Asamoto et al, 1990; Zhau et al, 1990). It has been suggested that amplification/overexpression of erbB-2 enhances cellular mitogenic signaling, thus accelerating tumorigenesis (Hynes et al, 1994). Despite the demonstrated correlation between erbB-2 overexpression and late-stage TCC, little is known about the tumor-initiating and -promoting potential of this gene in bladder epithelium.
c-myc: The c-myc proto-oncogene encodes a transcription factor that regulates cell proliferation and differentiation. Altered expression of c-myc gene is a common event in a variety of tumors including bladder cancer. Overexpression of c-myc protein, but not gene amplification, has been found in a significant number (up to 59%) of TCCs (Schmitz-Drager et al, 1997). However, the correlation of the c-myc overexpression with tumor stage is somewhat controversial, as different reports have shown either correlation with early (Masters et al, 1988; Sauter et al, 1995) or late (Kotake, 1990) stages or no correlation (Grimm et al, 1995; Kubota et al, 1995; Lipponen, 1995; Schmitz-Drager et al, 1997).
p53: A nuclear phosphoprotein, p53, plays a key role in cell cycle control, particularly in G1-S phase transition. Inactivating mutations of p53 can result in genome instability which predisposes cells to malignancy (Levine, 1997; Paulovich et al, 1997). The p53 gene can be inactivated by mutation or deletion of both alleles. It can also be functionally inactivated by a dominant-negative p53 mutant, which forms inactive hetero-oligomers with wild-type p53. In addition, it can be inactivated by oncogene products of DNA tumor viruses including SV40 large T antigen, adenoviral EIB protein, and papilloma viral protein EG (Vogelstein et al, 1992). p53 mutation is the most frequent genetic alteration in bladder cancer (50-60%). A number of studies have shown that p53 mutations occur in relatively late stages of human bladder cancer, suggesting that loss of p53 function is related to tumor invasion and progression (Reznikoff et al, 1996; Cordon-Cardo et al, 1997).
Rb: The functional inactivation of retinoblastoma gene (Rb) involving both alleles is found in around 30% of bladder cancer patients (Benedict, 1992). There is also a strong correlation between Rb inactivation and tumor stage. Cordon-Cardo and colleagues (1992) showed that Rb expression is altered in 38 of 48 muscle invasive bladder tumors, but only in 10 of 48 superficial tumors, suggesting that loss of Rb function contributes to bladder tumor progression.
p16: It has been suggested that chromosomal region 9p21 harbors a tumor suppressor gene because the loss of this region is identified in a significant number of bladder tumors. The search for this suppressor gene has led to the identification of the p16 gene. It is now known that p16 encodes a protein that inhibits the activity of cyclin-dependent kinase 4 and that is mutated in a wide variety of cancers (Kamb et al, 1994). p16-deficient mice are viable but highly prone to spontaneous and carcinogen-induced tumors (Serrano et al, 1996). In bladder cancer, the p16 mutation has been found in about 18% of the cases and is predominantly associated with low grade bladder tumors (Gruis et al, 1995; Orlow et al, 1995).
Although the genetic alterations involving the activation of oncogenes and inactivation of tumor suppressor genes are prevalent in human bladder cancer, their exact role in the multi-step bladder tumorigenesis has not been clearly defined. It remains unknown whether any of these alterations are responsible for the cellular transformation in normal urothelium and tumor progression, or some of them merely represent secondary events of well-advanced tumors (Adams et al, 1991; Fowlis et al, 1993).
SV40 large T antigen (SV40 Tag): As discussed above, p53 and Rb tumor suppressor genes frequently have loss-of-function mutations in human bladder cancer. SV40 Tag is an oncogene from DNA tumor virus, Simian Virus 40, SV40 that has been extensively used to study the role of p53 and Rd dysfunction in tumorigenesis. When introduced into host cells, SV40 Tag acts as a potent growth stimulator by inactivating p53 and Rb proteins, leading to uncontrolled cellular proliferation and tumor formation (Bryan et al, 1994).
Advances in recombinant DNA and genetic technologies have made it possible to introduce and express a desired recombinant gene sequence in a recipient animal. When only some of the animal's cells contain and express the introduced gene sequence while other cells do not, i.e., remain unaltered, the animal is known as a "chimeric" animal. The capacity of a chimeric animal to transmit the introduced gene sequence is present in the germ cells of the animal. Accordingly, only certain chimeric animals can pass along the desired gene sequence to their progeny.
By contrast, a "transgenic" animal contains the introduced gene sequence ("transgene") in all cells, and therefore, every transgenic animal is capable of transmitting the transgene to its progeny. The study of the molecular mechanism of tumorigenesis has been greatly facilitated in recent years by the use of a transgenic approach. This approach has made it possible to transfer genes of interest into a living animal which expresses the transgenes in desired tissues (Babinet et al, 1989). Transgenic animals containing and expressing oncogenes or tumor suppressor genes have been observed to be tumor-susceptible and/or to develop tumors (U.S. Pat. Nos. 4,736,866; 5,491,283; 5,550,316; 5,569,824; Stewart et al, 1984; Adams et al, 1985; Hanahan, 1985; Lacey et al, 1986; Palmiter et al, 1986; Cory et al, 1988). The specific effects or well-defined oncogenes or mutated tumor suppressor genes on cellular growth and differentiation can be assessed and correlated with multistage tumorigenesis under in vivo physiological conditions (Hanahan, 1989; Landel et al, 1990). The transgenic approach has also made it possible to study the cooperative activity of different oncogenes and mutated tumor suppressor genes by generating bi-transgenic animals that harbor two distinct genetic alterations (Berns, 1991; Fowlis et al, 1993). However, despite the clinical importance of bladder cancer, few animal models are available to study the molecular pathogenesis of bladder tumorigenesis. Although bladder tumors have been previously produced in rat urinary bladders using various chemical carcinogens, those tumors bear little similarity with those occurring in humans. Therefore, the value of these rat bladder tumors as a model system for human bladder cancers is limited.
Previous knowledge about the genetic alterations in bladder cancer formation largely derived from studies of well-advanced human bladder cancers. However, it is difficult to determine whether these alterations are the cause or the consequence of cancer development. Chemical carcinogenesis is capable of inducing bladder cancer formation, but the resulting genetic changes are often complex, and the resulting bladder phenotypes have little in common with the human counterpart.
Citation of any document herein is not intended as an admission that such document is pertinent prior art, or considered material to the patentability of any claim of the present application. Any statement as to content or a date of any document is based on the information available to applicant at the time of filing and does not constitute an admission as to the correctness of such a statement.